14 April 2015
Local and State Governments have progressively been undertaking flood mapping studies over the past two or three decades, with an initial emphasis on urban areas - understandably as these represent the majority of the Australian population base. Rural areas however, have been an on-going challenge for traditional flood modelling, given their vast floodplain areas, cross-catchment interactions and long flood event durations.
Current innovations in flood modelling techniques are now helping engineers to fill existing knowledge gaps regarding flood behaviour, especially in rural and remote locations. Light Detection And Ranging (LiDAR), also known as aerial laser survey, has transformed the collection of ground elevation data for flood modelling purposes. LiDAR enables an accurate three dimensional representation of the ground surface to be generated relatively rapidly. The resulting computerised surface can then be used directly in two-dimensional flood models.
Costs of LiDAR data collection are reducing due to rapid technological advancements in the size of equipment and in the carrier vehicle (e.g. gyroplanes, ultralights and unmanned aerial drones can now be fitted with LiDAR). For very large areas however, complete LiDAR coverage is still somewhat impractical. As an alternative, satellite data (e.g. NASA’s Shuttle Radar Topography Mission [SRTM] data) is being utilised with increasing success, recognising the limitations and compromise in accuracy associated with this very high-level data.
Another recent advancement is the development of rapid assessment computer models, such as the new TUFLOW-GPU software. By reducing some of the finer scale details typically required in contemporary two-dimensional models (such as one-dimensional linkages and complex structure configurations) and recompiling the software to run on parallelised Graphical Processing Units (GPUs) instead of Computational Processing Units (CPUs), simulation run-times have been reduced by a factor of up to 100 in many cases. Rapid assessment flood studies of this nature are particularly advantageous for applications where a high degree of accuracy is not warranted, such as:
It is anticipated that this rapid assessment flood modelling approach will, for the first time, enable whole floodplains of large river basins west of the Great Dividing Range to be modelled in two-dimensional space. It may also have strategic value for other areas that require a standardised and consistent basis for defining flood risks (e.g. for community-based flood mapping across whole LGAs that span multiple waterway catchments and floodplains).
BMT has been adopting the new rapid assessment TUFLOW-GPU software in combination with both LiDAR and SRTM data for a number of rural and remote projects over the past six months. Realisation of even “first pass” risk profiles for these areas has helped governments, insurance companies, emergency response agencies and the community better appreciate the risks associated with flooding, and empowering them to respond through appropriate floodplain risk management actions and planning.
The advancements in ground surface data collection and in flood modelling proficiency represent the dawn of a new era in floodplain management in Australia wherein flood risks can be assessed and managed at a holistic level (e.g. whole of river basin, or whole of jurisdictional area). Also, flood warning systems utilising such rapid modelling techniques can be linked to real-time data (e.g. rainfall, weather radar and upstream water levels) to better facilitate timely flood warnings, evacuation and emergency management demands.
BMT is continuing to utilise the rapid assessment TUFLOW-GPU system for a range of clients, and it is anticipated that the software will be available for commercial sale as an option within the existing TUFLOW suite in the first half of 2013
The potential opportunities and rewards are vast. Let’s change where we need to, let’s collaborate more. We will not just survive, but thrive!
Philip Haines discusses the partnership to develop a climate change adaptation plan in the Solomon Islands
Dr Darren Richardson, Ecology and Environmental Management Manager at BMT discusses the environmental issues and opportunities that must be addressed following a mine’s closure.
Freyja Lockwood at Bristol City Council discusses the emerging challenges of building trust and public acceptance of autonomous systems, artificial intelligence and data-driven innovation.